Warning, /swf-testbed/docs/diagram5_panda_provenance.md is written in an unsupported language. File is not indexed.
0001 # diagram5_panda_provenance
0002
0003 Scale-provenance stack. The "why 6 months is plausible" anchor: we're
0004 layering on a production WFMS with a decade of operational history,
0005 not starting from zero.
0006
0007 ```mermaid
0008 flowchart BT
0009 classDef app fill:#e3f2fd,stroke:#1565c0,stroke-width:2px,color:#000
0010 classDef ai fill:#f1f8e9,stroke:#33691e,stroke-width:2px,color:#000
0011 classDef panda fill:#fff3e0,stroke:#e65100,stroke-width:2px,color:#000
0012
0013 subgraph L1["<b>Foundation</b> — PanDA production WFMS (operational since 2005)"]
0014 direction LR
0015 P1["ATLAS @ LHC<br/>O(million) jobs/day<br/>200+ institutions"]:::panda
0016 P2["PanDA monitor<br/>deep drill-down<br/>refined 15+ years"]:::panda
0017 P3["ePIC production<br/>(monthly campaigns, OSG, HPC)"]:::panda
0018 P4["ePIC streaming<br/>workflow testbed<br/>(this team, 2025+)"]:::panda
0019 end
0020
0021 subgraph L2["AI instrumentation today — this team (2024–)"]
0022 direction LR
0023 I1["AskPanDA MCP"]:::ai
0024 I2["PanDA Monitor MCP"]:::ai
0025 I3["VectorDB RAG"]:::ai
0026 I4["Streaming Workflow MCP"]:::ai
0027 I5["3rd-party MCP (6+)"]:::ai
0028 end
0029
0030 subgraph L3["<b>New application layer</b> — LLM-driven orchestration (proposed, 6 months)"]
0031 direction LR
0032 A1["LLM workflow<br/>orchestrator"]:::app
0033 A2["Hybrid workflows<br/>LLM + deterministic"]:::app
0034 A3["Harnessed autonomous<br/>LLM action"]:::app
0035 A4["LLM research assistant<br/><i>evolution of Mattermost<br/>bot + codoc-ai</i>"]:::app
0036 end
0037
0038 L1 --> L2
0039 L2 --> L3
0040 ```
